Testing Framework Output¶

ParamSimMutation: Parameter(Simulation) - rate of mutation (evaluated at each speciation per gene)
ParamSimHGT: Parameter(Simulation) - rate of horizontal gene transfer (as percentage of speciated genome)
ParamSimDrop: Parameter(Simulation) - rate of gene drop (evaluated at each speciation per gene)
ParamTestDistance: Parameter(Testing) - Maximum ancestor distance from "root" species (average of all gene distances)
ParamTestTaxa: Parameter(Testing) - Minimum percent required of protein coverage for a species to be included next step
ParamTestProtein: Parameter(Testing) - Minimum percent required of ancestor coverage for a protein to be included in profile
ActualHost: Actual number of proteins present in host genome
ActualOriginHost: Actual number of proteins from host lineage present in host genome
ActualOriginSym: Actual number of proteins from symbiont lineage present in host genome
ActualHGT: Actual number of HGT events between symbiont and host
ResultHost: Calculated count of proteins in host profile
ResultSym: Calculated count of proteins in symbiont profile
ResultAmbig: Calculated count of proteins shared present in both host and symbiont profiles
ResultGoodHost: Calculated number of correct host profile identifications
ResultGoodSym: Calculated number of correct symbiont profile identifications
ResultAmbigHost: Calculated number of amibiguous host profile identifications (protein was also in symbiont profile)
ResultAmbigSym: Calculated number of ambiguous symiont profile identifications (protein was also in host profile)
ResultBadHost: Calculated number of incorrect host profile identifications
ResultBadSym: Calculated number of incorrect symbiont profile identifications
ResultHGTGood: Calculated number of correct HGT predictions
ResultHGTAmbig: Calculated number of corrent HGT predictions belonging to ambiguous profiles
ResultHGTBad: Calculated number of incorrect HGT predictions

The following is an example random sampling from the dataframe:

options(warn=-1) 
resultsSet <- read.csv("twestbrook-hgt1-results.txt", sep = "\t")
exampleSet <- resultsSet[sample(nrow(resultsSet), 5),]
cat(sprintf("Example set: %d out of %d rows\n", 5, nrow(resultsSet)))
tail(exampleSet)

Example set: 5 out of 360000 rows

Analysis¶

Effectiveness of phylogenetic profiling based on good profile hits minus bad hits out of total possible profile hits. Left doesn't remove ambiguous hits, while right does.

Given this, calculate the effectiveness for both host and symbiont, then plot to see correlation:

resultsSet$ResultEffectiveHost <- (resultsSet$ResultGoodHost - resultsSet$ResultBadHost) / resultsSet$ActualOriginHost
resultsSet$ResultEffectiveSym <- (resultsSet$ResultGoodSym - resultsSet$ResultBadSym) / resultsSet$ActualOriginSym
resultsSet$ResultEffectiveHostA <- (resultsSet$ResultGoodHost - resultsSet$ResultBadHost - resultsSet$ResultAmbigHost) / resultsSet$ActualOriginHost
resultsSet$ResultEffectiveSymA <- (resultsSet$ResultGoodSym - resultsSet$ResultBadSym  - resultsSet$ResultAmbigSym) / resultsSet$ActualOriginSym

# Normalize from -1:1 to 0:1
resultsSet$ResultEffectiveHost <- (resultsSet$ResultEffectiveHost + 1) / 2
resultsSet$ResultEffectiveSym <- (resultsSet$ResultEffectiveSym + 1) / 2
resultsSet$ResultEffectiveHostA <- (resultsSet$ResultEffectiveHostA + 1) / 2
resultsSet$ResultEffectiveSymA <- (resultsSet$ResultEffectiveSymA + 1) / 2

library(gridExtra)
library(ggplot2)
options(repr.plot.width = 15, repr.plot.height = 15)

p1 <- ggplot(resultsSet, aes(ResultEffectiveHost, ResultEffectiveSym, color=ParamSimMutation)) + geom_point(alpha = 1/20) + scale_colour_gradient(low="black", high="green") 
p2 <- ggplot(resultsSet, aes(ResultEffectiveHostA, ResultEffectiveSymA, color=ParamSimMutation)) + geom_point(alpha = 1/20)  + scale_colour_gradient(low="black", high="green")
p3 <- ggplot(resultsSet, aes(ResultEffectiveHost, ResultEffectiveSym, color=ParamSimHGT)) + geom_point(alpha = 1/20) + scale_colour_gradient(low="black", high="green") 
p4 <- ggplot(resultsSet, aes(ResultEffectiveHostA, ResultEffectiveSymA, color=ParamSimHGT)) + geom_point(alpha = 1/20)  + scale_colour_gradient(low="black", high="green")
p5 <- ggplot(resultsSet, aes(ResultEffectiveHost, ResultEffectiveSym, color=ParamSimDrop)) + geom_point(alpha = 1/20) + scale_colour_gradient(low="black", high="green") 
p6 <- ggplot(resultsSet, aes(ResultEffectiveHostA, ResultEffectiveSymA, color=ParamSimDrop)) + geom_point(alpha = 1/20)  + scale_colour_gradient(low="black", high="green")

grid.arrange(p1, p2, p3, p4, p5, p6,  ncol = 2, nrow = 3)

p1 <- ggplot(resultsSet, aes(ResultEffectiveHost, ResultEffectiveSym, color=ParamTestDistance)) + geom_point(alpha = 1/20) + scale_colour_gradient(low="black", high="green") 
p2 <- ggplot(resultsSet, aes(ResultEffectiveHostA, ResultEffectiveSymA, color=ParamTestDistance)) + geom_point(alpha = 1/20)  + scale_colour_gradient(low="black", high="green")
p3 <- ggplot(resultsSet, aes(ResultEffectiveHost, ResultEffectiveSym, color=ParamTestTaxa)) + geom_point(alpha = 1/20) + scale_colour_gradient(low="black", high="green") 
p4 <- ggplot(resultsSet, aes(ResultEffectiveHostA, ResultEffectiveSymA, color=ParamTestTaxa)) + geom_point(alpha = 1/20)  + scale_colour_gradient(low="black", high="green")
p5 <- ggplot(resultsSet, aes(ResultEffectiveHost, ResultEffectiveSym, color=ParamTestProtein)) + geom_point(alpha = 1/20) + scale_colour_gradient(low="black", high="green") 
p6 <- ggplot(resultsSet, aes(ResultEffectiveHostA, ResultEffectiveSymA, color=ParamTestProtein)) + geom_point(alpha = 1/20)  + scale_colour_gradient(low="black", high="green")

grid.arrange(p1, p2, p3, p4, p5, p6,  ncol = 2, nrow = 3)

Look at the correlation across specific simulation combinations.

First row: All three rates at 0.1, 0.3, and 0.5 percent

Each subsequent row - Two rates at 0.1, 0.3, and 0.5 with the third rate fixed at 0.1. (In order: Mutation, HGT, then Drop)

options(repr.plot.width = 10, repr.plot.height = 3)
p1 <- ggplot(resultsSet[resultsSet$ParamSimMutation == 0.1 & resultsSet$ParamSimHGT == 0.1 & resultsSet$ParamSimDrop == 0.1,], aes(ResultEffectiveHostA, ResultEffectiveSymA)) + geom_point(alpha = 1/10)
p2 <- ggplot(resultsSet[resultsSet$ParamSimMutation == 0.3 & resultsSet$ParamSimHGT == 0.3 & resultsSet$ParamSimDrop == 0.3,], aes(ResultEffectiveHostA, ResultEffectiveSymA)) + geom_point(alpha = 1/10)
p3 <- ggplot(resultsSet[resultsSet$ParamSimMutation == 0.5 & resultsSet$ParamSimHGT == 0.5 & resultsSet$ParamSimDrop == 0.5,], aes(ResultEffectiveHostA, ResultEffectiveSymA)) + geom_point(alpha = 1/10)

grid.arrange(p1, p2, p3, ncol = 3, nrow = 1)#, widths=c(1,1,1))

Mutation rate fixed at 0.1

p1 <- ggplot(resultsSet[resultsSet$ParamSimMutation == 0.1 & resultsSet$ParamSimHGT == 0.1 & resultsSet$ParamSimDrop == 0.1,], aes(ResultEffectiveHostA, ResultEffectiveSymA)) + geom_point(alpha = 1/10)
p2 <- ggplot(resultsSet[resultsSet$ParamSimMutation == 0.1 & resultsSet$ParamSimHGT == 0.3 & resultsSet$ParamSimDrop == 0.3,], aes(ResultEffectiveHostA, ResultEffectiveSymA)) + geom_point(alpha = 1/10)
p3 <- ggplot(resultsSet[resultsSet$ParamSimMutation == 0.1 & resultsSet$ParamSimHGT == 0.5 & resultsSet$ParamSimDrop == 0.5,], aes(ResultEffectiveHostA, ResultEffectiveSymA)) + geom_point(alpha = 1/10)

grid.arrange(p1, p2, p3, ncol = 3, nrow = 1)

HGT rate fixed at 0.1

p1 <- ggplot(resultsSet[resultsSet$ParamSimMutation == 0.1 & resultsSet$ParamSimHGT == 0.1 & resultsSet$ParamSimDrop == 0.1,], aes(ResultEffectiveHostA, ResultEffectiveSymA)) + geom_point(alpha = 1/10)
p2 <- ggplot(resultsSet[resultsSet$ParamSimMutation == 0.3 & resultsSet$ParamSimHGT == 0.1 & resultsSet$ParamSimDrop == 0.3,], aes(ResultEffectiveHostA, ResultEffectiveSymA)) + geom_point(alpha = 1/10)
p3 <- ggplot(resultsSet[resultsSet$ParamSimMutation == 0.5 & resultsSet$ParamSimHGT == 0.1 & resultsSet$ParamSimDrop == 0.5,], aes(ResultEffectiveHostA, ResultEffectiveSymA)) + geom_point(alpha = 1/10)

grid.arrange(p1, p2, p3, ncol = 3, nrow = 1)

Drop rate fixed at 0.1

p1 <- ggplot(resultsSet[resultsSet$ParamSimMutation == 0.1 & resultsSet$ParamSimHGT == 0.1 & resultsSet$ParamSimDrop == 0.1,], aes(ResultEffectiveHostA, ResultEffectiveSymA)) + geom_point(alpha = 1/10)
p2 <- ggplot(resultsSet[resultsSet$ParamSimMutation == 0.3 & resultsSet$ParamSimHGT == 0.3 & resultsSet$ParamSimDrop == 0.1,], aes(ResultEffectiveHostA, ResultEffectiveSymA)) + geom_point(alpha = 1/10)
p3 <- ggplot(resultsSet[resultsSet$ParamSimMutation == 0.5 & resultsSet$ParamSimHGT == 0.5 & resultsSet$ParamSimDrop == 0.1,], aes(ResultEffectiveHostA, ResultEffectiveSymA)) + geom_point(alpha = 1/10)

grid.arrange(p1, p2, p3, ncol = 3, nrow = 1)

Now compensate for ambiguous hits as well, as view effect

Accounting for ambiguous cases seems to add a senario (given some simulation/testing parameters) that causes a negative correlation. Investigate this

Calculate effective HGT by taking good HGT hits minux bad HGT hits out of all possible HGT events. Then plot in relation to both the host profile and the symbiont profile

resultsSet$ResultEffectiveHGT <- (resultsSet$ResultHGTGood - resultsSet$ResultHGTBad - resultsSet$ResultHGTAmbig) / resultsSet$ActualHGT
resultsSet$ResultEffectiveHGT <- (resultsSet$ResultEffectiveHGT + 1) / 2

options(repr.plot.width = 15, repr.plot.height = 15)
p1 <- ggplot(resultsSet, aes(ResultEffectiveHostA, ResultEffectiveHGT, color=ParamSimMutation)) + geom_point(alpha = 1/10) + scale_colour_gradient(low="black", high="green") 
p2 <- ggplot(resultsSet, aes(ResultEffectiveSymA, ResultEffectiveHGT, color=ParamSimMutation)) + geom_point(alpha = 1/10) + scale_colour_gradient(low="black", high="green") 
p3 <- ggplot(resultsSet, aes(ResultEffectiveHostA, ResultEffectiveHGT, color=ParamSimHGT)) + geom_point(alpha = 1/10) + scale_colour_gradient(low="black", high="green") 
p4 <- ggplot(resultsSet, aes(ResultEffectiveSymA, ResultEffectiveHGT, color=ParamSimHGT)) + geom_point(alpha = 1/10) + scale_colour_gradient(low="black", high="green") 
p5 <- ggplot(resultsSet, aes(ResultEffectiveHostA, ResultEffectiveHGT, color=ParamSimDrop)) + geom_point(alpha = 1/10) + scale_colour_gradient(low="black", high="green") 
p6 <- ggplot(resultsSet, aes(ResultEffectiveSymA, ResultEffectiveHGT, color=ParamSimDrop)) + geom_point(alpha = 1/10) + scale_colour_gradient(low="black", high="green") 

grid.arrange(p1, p2, p3, p4, p5, p6,  ncol = 2, nrow = 3)

p1 <- ggplot(resultsSet, aes(ResultEffectiveHostA, ResultEffectiveHGT, color=ParamTestDistance)) + geom_point(alpha = 1/10) + scale_colour_gradient(low="black", high="green") 
p2 <- ggplot(resultsSet, aes(ResultEffectiveSymA, ResultEffectiveHGT, color=ParamTestDistance)) + geom_point(alpha = 1/10) + scale_colour_gradient(low="black", high="green") 
p3 <- ggplot(resultsSet, aes(ResultEffectiveHostA, ResultEffectiveHGT, color=ParamTestTaxa)) + geom_point(alpha = 1/10) + scale_colour_gradient(low="black", high="green") 
p4 <- ggplot(resultsSet, aes(ResultEffectiveSymA, ResultEffectiveHGT, color=ParamTestTaxa)) + geom_point(alpha = 1/10) + scale_colour_gradient(low="black", high="green") 
p5 <- ggplot(resultsSet, aes(ResultEffectiveHostA, ResultEffectiveHGT, color=ParamTestProtein)) + geom_point(alpha = 1/10) + scale_colour_gradient(low="black", high="green") 
p6 <- ggplot(resultsSet, aes(ResultEffectiveSymA, ResultEffectiveHGT, color=ParamTestProtein)) + geom_point(alpha = 1/10) + scale_colour_gradient(low="black", high="green") 

grid.arrange(p1, p2, p3, p4, p5, p6,  ncol = 2, nrow = 3)

Condense effectiveness into a single value by averaging these two values. Isolate the parameters and this total effectiveness into a new dataframe. Then show a correlation matrix, and pairwise comparisons for simulation and testing parameters against total effectiveness.

parameterSet <- data.frame(resultsSet$ParamSimMutation, 
                           resultsSet$ParamSimHGT, 
                           resultsSet$ParamSimDrop, 
                           resultsSet$ParamTestDistance, 
                           resultsSet$ParamTestTaxa, 
                           resultsSet$ParamTestProtein,
                           resultsSet$ResultEffectiveHostA)
cor(parameterSet, use="complete")

parameterSet <- data.frame(resultsSet$ParamSimMutation, 
                           resultsSet$ParamSimHGT, 
                           resultsSet$ParamSimDrop, 
                           resultsSet$ParamTestDistance, 
                           resultsSet$ParamTestTaxa, 
                           resultsSet$ParamTestProtein,
                           resultsSet$ResultEffectiveSymA)
cor(parameterSet, use="complete")

resultsSet$ResultEffectiveCombined <- (resultsSet$ResultEffectiveHostA + resultsSet$ResultEffectiveSymA) / 2
parameterSet <- data.frame(resultsSet$ParamSimMutation, 
                           resultsSet$ParamSimHGT, 
                           resultsSet$ParamSimDrop, 
                           resultsSet$ParamTestDistance, 
                           resultsSet$ParamTestTaxa, 
                           resultsSet$ParamTestProtein,
                           resultsSet$ResultEffectiveCombined)
cor(parameterSet, use="complete")

Testing to see distributions of total effectiveness by parameter type

p1 <- ggplot(resultsSet, aes(ParamSimMutation, ResultEffectiveCombined, group=cut_width(ParamSimMutation, 0.10))) + geom_boxplot(outlier.color=alpha("black", 1/100))
p2 <- ggplot(resultsSet, aes(ParamSimHGT, ResultEffectiveCombined, group=cut_width(ParamSimHGT, 0.10))) + geom_boxplot(outlier.color=alpha("black", 1/100))
p3 <- ggplot(resultsSet, aes(ParamSimDrop, ResultEffectiveCombined, group=cut_width(ParamSimDrop, 0.10))) + geom_boxplot(outlier.color=alpha("black", 1/100))
p4 <- ggplot(resultsSet, aes(ParamTestDistance, ResultEffectiveCombined, group=cut_width(ParamTestDistance, 0.10))) + geom_boxplot(outlier.color=alpha("black", 1/100))
p5 <- ggplot(resultsSet, aes(ParamTestTaxa, ResultEffectiveCombined, group=cut_width(ParamTestTaxa, 0.10))) + geom_boxplot(outlier.color=alpha("black", 1/100))
p6 <- ggplot(resultsSet, aes(ParamTestProtein, ResultEffectiveCombined, group=cut_width(ParamTestProtein, 0.10))) + geom_boxplot(outlier.color=alpha("black", 1/100))

options(repr.plot.width = 10, repr.plot.height = 6)
grid.arrange(p1, p2, p3, p4, p5, p6, nrow=2, ncol=3)

chartList = sapply(c(0.2, 0.4, 0.6, 0.8), function(activeTaxa) {
    rowList = lapply(c(0.2, 0.4, 0.6, 0.8), function (activeProtein) {
        resultsSubset <- resultsSet[resultsSet$ParamTestTaxa == activeTaxa & resultsSet$ParamTestProtein == activeProtein & resultsSet$ParamSimDrop == 0.1 & resultsSet$ParamTestDistance == 0.7,]
        effectiveColors <- rgb(resultsSubset$ResultEffectiveHostA, ifelse(is.nan(resultsSubset$ResultEffectiveSymA),0,resultsSubset$ResultEffectiveSymA), 0)
        #effectiveColors <- hsv(resultsSubset$ResultEffectiveHostA, ifelse(is.nan(resultsSubset$ResultEffectiveSymA),0,resultsSubset$ResultEffectiveSymA) , 0.5 )
        hgtColors <- rgb(0, 0, ifelse(is.nan(resultsSubset$ResultHGTAmbig/resultsSubset$ActualHGT), 0, resultsSubset$ResultHGTAmbig/resultsSubset$ActualHGT))
        activePlot <- ggplot(resultsSubset, aes(x=ParamSimMutation, y=ParamSimHGT)) + 
            geom_tile(color="white", fill=effectiveColors) +
            geom_point(size=resultsSubset$ResultEffectiveHGT * 4, shape=16, color="blue", alpha=0.5) +
            theme(axis.title=element_text(size=8))

        if (activeProtein != 0.2) activePlot <- activePlot + theme(axis.title.y=element_blank(), axis.text.y=element_blank(), axis.ticks.y=element_blank(),)
        if (activeTaxa != 0.8) activePlot <- activePlot + theme(axis.title.x=element_blank(), axis.text.x=element_blank(), axis.ticks.x=element_blank(),)
        activePlot
            
    })
})     
      
options(repr.plot.width = 10, repr.plot.height = 6)
do.call("grid.arrange", c(chartList, nrow=4, ncol=4, left = "Testing: Taxa", bottom = "Testing: Protein"))

library(reshape2)
resultsSubset1 <- resultsSet[resultsSet$ResultEffectiveHostA > 0.80 & resultsSet$ResultEffectiveSymA > 0.80 & resultsSet$ResultEffectiveHGT > 0.80,]
focusSubset <- data.frame(resultsSubset1$ParamTestDistance, resultsSubset1$ParamTestTaxa, resultsSubset1$ParamTestProtein)
options(repr.plot.width = 4, repr.plot.height = 4)
ggplot(melt(focusSubset), aes(x = variable, y = value)) + geom_boxplot() + 
    scale_x_discrete(labels=c("Distance","Taxa","Protein"))  + 
    scale_y_continuous(breaks = pretty(resultsSubset1$ParamTestDistance, n = 10))

No id variables; using all as measure variables

resultsBaseSubset <- resultsSet[resultsSet$ParamTestDistance == 0.6 & resultsSet$ParamSimMutation <= 0.2 & resultsSet$ParamSimHGT > 0 & resultsSet$ParamSimHGT <= 0.3,]
chartList = lapply(c(0.4, 0.5, 0.6), function(activeTaxa) {
    resultsSubset1 <- resultsBaseSubset[resultsSet$ParamTestTaxa == activeTaxa & resultsSet$ParamTestProtein == 0.4,]
    resultsSubset2 <- resultsBaseSubset[resultsSet$ParamTestTaxa == activeTaxa & resultsSet$ParamTestProtein == 0.5,]
    resultsSubset3 <- resultsBaseSubset[resultsSet$ParamTestTaxa == activeTaxa & resultsSet$ParamTestProtein == 0.6,]

    activePlot <- ggplot() + 
        geom_point(data=resultsSubset1, aes(ResultEffectiveHostA, ResultEffectiveHGT, color="0.4"), alpha=1/3) + 
        geom_point(data=resultsSubset2, aes(ResultEffectiveHostA, ResultEffectiveHGT, color="0.5"), alpha=1/3) +
        geom_point(data=resultsSubset3, aes(ResultEffectiveHostA, ResultEffectiveHGT, color="0.6"), alpha=1/3) +
        stat_smooth(data=resultsSubset1, aes(ResultEffectiveHostA, ResultEffectiveHGT), method = "loess", formula = y ~ x, size = 1, col = "red") +
        stat_smooth(data=resultsSubset2, aes(ResultEffectiveHostA, ResultEffectiveHGT), method = "loess", formula = y ~ x, size = 1, col = "green") +
        stat_smooth(data=resultsSubset3, aes(ResultEffectiveHostA, ResultEffectiveHGT), method = "loess", formula = y ~ x, size = 1, col = "blue") +
        coord_cartesian(xlim = c(0.5, 1), ylim = c(0, 1)) +
        scale_color_manual("Protein", values=c("red", "green", "blue"))
    
    if (activeTaxa != 0.4) activePlot <- activePlot + theme(axis.title.y=element_blank(), axis.text.y=element_blank(), axis.ticks.y=element_blank(),)
    if (activeTaxa != 0.6) activePlot <- activePlot + theme(legend.position="none")
    
    activePlot
})     
    
options(repr.plot.width = 15, repr.plot.height = 5)
do.call("grid.arrange", c(chartList, nrow=1, ncol=3,  bottom = "Testing: Taxa"))

	ParamSimMutation	ParamSimHGT	ParamSimDrop	ParamTestDistance	ParamTestTaxa	ParamTestProtein	ActualHost	ActualOriginHost	ActualOriginSym	ActualHGT	⋯	ResultAmbig	ResultGoodHost	ResultGoodSym	ResultAmbigHost	ResultAmbigSym	ResultBadHost	ResultBadSym	ResultHGTGood	ResultHGTAmbig	ResultHGTBad
79688	0.2	0.1	0.1	0.6	0.8	0.7	523	420	52	10	⋯	25	357	36	7	15	63	16	2	0	8
20206	0.0	0.3	0.2	0.2	0.0	0.5	705	313	181	63	⋯	376	148	101	148	101	165	80	28	28	35
10739	0.0	0.1	0.4	0.7	0.3	0.8	260	184	39	22	⋯	0	149	8	0	0	35	31	2	0	20
232043	0.6	0.2	0.4	0.0	0.4	0.2	341	170	98	61	⋯	0	0	0	0	0	170	98	0	0	61
126219	0.3	0.3	0.0	0.2	0.1	0.8	978	500	233	0	⋯	0	0	0	0	0	500	233	0	0	0

	resultsSet.ParamSimMutation	resultsSet.ParamSimHGT	resultsSet.ParamSimDrop	resultsSet.ParamTestDistance	resultsSet.ParamTestTaxa	resultsSet.ParamTestProtein	resultsSet.ResultEffectiveHostA
resultsSet.ParamSimMutation	1.000000e+00	-1.524576e-18	-7.449990e-19	4.248567e-19	-2.632932e-21	-3.285005e-21	-0.43997114
resultsSet.ParamSimHGT	-1.524576e-18	1.000000e+00	0.000000e+00	5.176879e-19	0.000000e+00	0.000000e+00	-0.23056593
resultsSet.ParamSimDrop	-7.449990e-19	0.000000e+00	1.000000e+00	5.127762e-19	0.000000e+00	0.000000e+00	-0.02683095
resultsSet.ParamTestDistance	4.248567e-19	5.176879e-19	5.127762e-19	1.000000e+00	-1.916519e-23	-3.285005e-21	0.38786393
resultsSet.ParamTestTaxa	-2.632932e-21	0.000000e+00	0.000000e+00	-1.916519e-23	1.000000e+00	-6.570466e-21	-0.05015466
resultsSet.ParamTestProtein	-3.285005e-21	0.000000e+00	0.000000e+00	-3.285005e-21	-6.570466e-21	1.000000e+00	-0.07280770
resultsSet.ResultEffectiveHostA	-4.399711e-01	-2.305659e-01	-2.683095e-02	3.878639e-01	-5.015466e-02	-7.280770e-02	1.00000000

	resultsSet.ParamSimMutation	resultsSet.ParamSimHGT	resultsSet.ParamSimDrop	resultsSet.ParamTestDistance	resultsSet.ParamTestTaxa	resultsSet.ParamTestProtein	resultsSet.ResultEffectiveSymA
resultsSet.ParamSimMutation	1.000000e+00	-4.651744e-17	-2.861173e-18	-5.414363e-19	3.682235e-20	-5.745176e-21	-0.50656907
resultsSet.ParamSimHGT	-4.651744e-17	1.000000e+00	3.993435e-18	1.064770e-18	-6.406336e-20	5.560669e-25	-0.11342481
resultsSet.ParamSimDrop	-2.861173e-18	3.993435e-18	1.000000e+00	5.127762e-19	0.000000e+00	0.000000e+00	-0.08318546
resultsSet.ParamTestDistance	-5.414363e-19	1.064770e-18	5.127762e-19	1.000000e+00	-2.299823e-23	-3.284914e-21	0.22437817
resultsSet.ParamTestTaxa	3.682235e-20	-6.406336e-20	0.000000e+00	-2.299823e-23	1.000000e+00	-6.570375e-21	-0.09494226
resultsSet.ParamTestProtein	-5.745176e-21	5.560669e-25	0.000000e+00	-3.284914e-21	-6.570375e-21	1.000000e+00	-0.09237689
resultsSet.ResultEffectiveSymA	-5.065691e-01	-1.134248e-01	-8.318546e-02	2.243782e-01	-9.494226e-02	-9.237689e-02	1.00000000

	resultsSet.ParamSimMutation	resultsSet.ParamSimHGT	resultsSet.ParamSimDrop	resultsSet.ParamTestDistance	resultsSet.ParamTestTaxa	resultsSet.ParamTestProtein	resultsSet.ResultEffectiveCombined
resultsSet.ParamSimMutation	1.000000e+00	-4.651744e-17	-2.861173e-18	-5.414363e-19	3.682235e-20	-5.745176e-21	-0.47198411
resultsSet.ParamSimHGT	-4.651744e-17	1.000000e+00	3.993435e-18	1.064770e-18	-6.406336e-20	5.560669e-25	-0.15931000
resultsSet.ParamSimDrop	-2.861173e-18	3.993435e-18	1.000000e+00	5.127762e-19	0.000000e+00	0.000000e+00	-0.05660427
resultsSet.ParamTestDistance	-5.414363e-19	1.064770e-18	5.127762e-19	1.000000e+00	-2.299823e-23	-3.284914e-21	0.32665964
resultsSet.ParamTestTaxa	3.682235e-20	-6.406336e-20	0.000000e+00	-2.299823e-23	1.000000e+00	-6.570375e-21	-0.08957690
resultsSet.ParamTestProtein	-5.745176e-21	5.560669e-25	0.000000e+00	-3.284914e-21	-6.570375e-21	1.000000e+00	-0.08543826
resultsSet.ResultEffectiveCombined	-4.719841e-01	-1.593100e-01	-5.660427e-02	3.266596e-01	-8.957690e-02	-8.543826e-02	1.00000000